Research On Construction And Application Of Knowledge Graph Of Health Domain For Traditional Chinese Medicine Syndrome

In recent years,with the popularization of the concept of health management,people have paid more and more attention to the health care of the body.The idea of "preventive treatment of diseases " of traditional Chinese medicine(TCM)adopts the concept and method of early intervention,which can effectively achieve the goal of maintaining health and preventing disease development.This fits well with the concept of health management.However,the relevant knowledge of TCM is more complicated and has certain domain characteristics.Integrating multi-source TCM knowledge,exploring the internal relationship between knowledge and providing convenient knowledge retrieval services have important research significance for TCM theory to guide health management.In the face of the above problems,the knowledge graph is an effective solution.As a networked knowledge system,the knowledge graph can contain a large amount of semantic information,capture and present the semantic relationship between domain concepts,and realize the connection of trivial and scattered knowledge.Because of the good performance on the knowledge retrieval,automatic question and answering,it has been widely used in many fields such as business and medical.Knowledge graph has also made some progress in the field of traditional Chinese medicine,such as the Chinese medicine knowledge service platform based on the knowledge graph which is constructed by China Academy of Traditional Chinese Medicine,but most of them are used in clinical direction,and the semantic information in traditional Chinese medicine has not been fully mined.Applying the knowledge graph to the field of Chinese medicine health can effectively manage related entities such as syndromes,symptoms,diseases,etc.,and can help users to obtain health management knowledge more conveniently.Using knowledge graph technology to extract and integrate TCM knowledge is an important direction for the construction of knowledge graph in the health domain,and it is also our research interest.This paper studies the specific application of knowledge graph in the field of traditional Chinese medicine and the related progress of knowledge graph construction technology.On this basis,the following work is carried out: Firstly,the paper completes the construction of the knowledge graph of Chinese medicine health domain.The core entities in the knowledge graph are defined,including four kinds of entities: syndrome,symptom,disease,and treatment.Then,the semantic relationship between the four entities is defined based on the characteristics of the entity.At last,complete the definition of data schema by constructing domain ontology.Secondly,for the named entity recognition,this paper proposes an entity recognition algorithm based on word vector splicing.It achieves a better result on entity recognition by combining the domain dictionary.Thirdly,this paper proposes a semantic retrieval model based on the knowledge graph.We study semantic search and propose a semantic retrieval model for TCM health management.Compared with the traditional keyword-based information retrieval method,this model can better understand the user's search intention.Fourthly,this paper constructs a health management platform based on the knowledge graph of Chinese medicine health domain.The platform mainly composes of two parts: on the one hand,it provides functions such as concept management and entity management for domain-oriented experts,which can extend the data model and entity of knowledge graph;on the other hand,it provides users with knowledge retrieval and automatic question and answering services,which can provide support for users' daily health management.